curl --request POST \
  --url https://aireiter.com/api/openapi/submit \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "gpt_image_2_official",
    "params": {
      "prompt": "Ancient castle under the starry sky",
      "aspect_ratio": "16:9",
      "resolution": "2K",
      "quality": "high"
    },
    "out_task_id": "my_task_123456"
  }'

{
  "code": 200,
  "data": [
    {
      "status": "submitted",
      "task_id": "task_01KPTXXXXXXXXXXXXXXX"
    }
  ]
}

GPT-Image-2

GPT-Image-2 Official Channel Image Generation

Official OpenAI gpt-image-2 model, compatible with the /v1/images/generations protocol
Asynchronous processing mode, returns task_id for subsequent queries
Text-to-image / Image-to-image all-in-one
Added resolution tier field, supports 1K / 2K / 4K resolution options
Supports 13 aspect ratios (4K tier supports 6 of them: 16:9 / 9:16 / 2:1 / 1:2 / 21:9 / 9:21)
Up to 9 reference images, supports mixed URL and base64
Submitted prompts undergo platform sensitive words / safety review, violating content will be directly rejected

POST

api

openapi

submit

curl --request POST \
  --url https://aireiter.com/api/openapi/submit \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "gpt_image_2_official",
    "params": {
      "prompt": "Ancient castle under the starry sky",
      "aspect_ratio": "16:9",
      "resolution": "2K",
      "quality": "high"
    },
    "out_task_id": "my_task_123456"
  }'

{
  "code": 200,
  "data": [
    {
      "status": "submitted",
      "task_id": "task_01KPTXXXXXXXXXXXXXXX"
    }
  ]
}

Authorizations

Authorization

string

required

All endpoints require authentication using a Bearer Token.Get your API Key:Visit the API Key Management Page to obtain your API Key.Add the following to the request header when using:

Authorization: Bearer YOUR_API_KEY

Body

model

string

default:"gpt_image_2_official"

required

Name of the image generation modelFixed value gpt_image_2_official (OpenAI official gpt-image-2 model)

params

object

required

Model parameters object

Show params object properties

prompt

string

required

Text description for image generation

Supports Chinese and English, detailed description recommended
Will be checked by platform sensitive word / security review before submitting; violation will cause an error response

aspect_ratio

string

default:"1:1"

Aspect ratioExternally used ratio values; internally mapped to specific pixels according to resolution.Supports 13 ratios:

1:1 - Square composition (default, social avatar / logo)
3:2 - Landscape composition (common DSLR ratio)
2:3 - Portrait composition (poster vertical)
4:3 - Landscape composition (classic monitor / PPT)
3:4 - Portrait composition
5:4 - Landscape composition
4:5 - Portrait composition (Instagram vertical post)
16:9 - Landscape composition (widescreen video cover)
9:16 - Portrait composition (full-screen phone / short video cover)
2:1 - Landscape composition (web banner)
1:2 - Portrait composition
21:9 - Landscape composition (cinema widescreen)
9:21 - Portrait composition

Only ratio format is supported; pixel size inputs (e.g. 1024x1024) will cause an error

resolution

string

default:"1K"

Resolution tierControls actual image clarity:

1K - 1024 baseline, cost-effective for daily use (default)
2K - 2048 baseline, suitable for posters / HD needs
4K - 3840 baseline, supported only by 6 ratios (16:9 / 9:16 / 2:1 / 1:2 / 21:9 / 9:21)

Unsupported 4K combinations will return 400 error:

1:1 × 4K ❌ (exceeds pixel limit)
3:2 / 2:3 × 4K ❌ (exceeds pixel limit)
4:3 / 3:4 × 4K ❌ (exceeds pixel limit)
5:4 / 4:5 × 4K ❌ (exceeds pixel limit)

For HD images in these ratios, use resolution=2K instead.

quality

string

default:"low"

Image quality

low - Fast and cost-saving, contour adequate (default)
medium - Balanced
high - Highest precision (4K + high consumes >120 seconds)

image_url

array

Reference image array, uses image-to-image mode when provided

Up to 9 reference images
Supports two formats:
1. Publicly accessible URL
2. Base64 encoded format
  - Must use full Data URI format: data:image/{format};base64,{base64 data}
  - Supported image formats: jpeg, png, webp
  - Example: data:image/jpeg;base64,/9j/4AAQSkZJRgABAQEAYABg...
  - ⚠️ Note: must include the data:image/jpeg;base64, prefix

out_task_id

string

required

Initiator task IDUser-defined task identifier, required

Size × Resolution Mapping Table

aspect_ratio × resolution → OpenAI Actual Pixels (13 ratios × 3 tiers):

aspect_ratio	`1K`	`2K`	`4K`
`1:1`	1024×1024	2048×2048	❌ Pixel limit exceeded
`3:2`	1536×1024	2048×1360	❌ Pixel limit exceeded
`2:3`	1024×1536	1360×2048	❌ Pixel limit exceeded
`4:3`	1024×768	2048×1536	❌ Pixel limit exceeded
`3:4`	768×1024	1536×2048	❌ Pixel limit exceeded
`5:4`	1280×1024	2560×2048	❌ Pixel limit exceeded
`4:5`	1024×1280	2048×2560	❌ Pixel limit exceeded
`16:9`	1536×864	2048×1152	3840×2160
`9:16`	864×1536	1152×2048	2160×3840
`2:1`	2048×1024	2688×1344	3840×1920
`1:2`	1024×2048	1344×2688	1920×3840
`21:9`	2016×864	2688×1152	3840×1648
`9:21`	864×2016	1152×2688	1648×3840

3:2 / 2:3 @ 2K are actually 2048×1360 (approximate ratio, error < 0.5%); 4K supports only 6 ratios (combinations exceeding OpenAI’s total pixel limit are unavailable).

Response

code

integer

Response status code

data

array

Returned data array

Show Properties

status

string

Task status

submitted - Submitted

task_id

string

Unique task identifier, used for subsequent querying of the task result

Query Task Result

Upon successful submission, a task_id is returned. Poll the task status using GET /v1/tasks/{task_id}. For details, see the Task Query Endpoint.

Polling Recommendations

Initial Query Delay: Wait 10~20 seconds after submission before starting to query
Query Interval: Recommend every 3~5 seconds
Timeout Reference: high + 2K/4K combination can take up to 130 seconds; client timeout recommendation ≥ 180 seconds

curl --request POST \
  --url https://aireiter.com/api/openapi/submit \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
    "model": "gpt_image_2_official",
    "params": {
      "prompt": "Ancient castle under the starry sky",
      "aspect_ratio": "16:9",
      "resolution": "2K",
      "quality": "high"
    },
    "out_task_id": "my_task_123456"
  }'

{
  "code": 200,
  "data": [
    {
      "status": "submitted",
      "task_id": "task_01KPTXXXXXXXXXXXXXXX"
    }
  ]
}

GPT-Image-2 Image Generation Seedream V4 Image Generation

Documentation Index

​Authorizations

​Body

​Size × Resolution Mapping Table

​Response

​Query Task Result

​Polling Recommendations

Authorizations

Body

Size × Resolution Mapping Table

Response

Query Task Result

Polling Recommendations